Acute kidney injury (AKI) is a devastating complication of acute illness that affects more than 500,000 hospitalized patients annually in the US alone. AKI markedly increases risk of in-hospital mortality, and confers greater risk of developing incident chronic kidney disease (CKD) and more rapid progression of established CKD. Therapeutic advances to improve outcomes in AKI have been impeded by lack of sufficiently sensitive diagnostics tests to detect early kidney injury before serum creatinine rise, and by a dearth of actionable pathogenic targets for clinical intervention. Novel diagnostic markers of early AKI and novel therapeutic targets are desperately needed to improve outcomes. Circulating levels of the phosphate-regulating hormone, fibroblast growth factor 23 (FGF23), increase progressively in patients with CKD, beginning early in the course of disease when kidney function is only subtly impaired. This exquisite sensitivity to even mild kidney dysfunction suggests that FGF23 could be an ideal biomarker for early injury in AKI. Our recently published study of rodent models of AKI supports this paradigm: FGF23 levels rise sharply soon after inducing AKI and before changes can be detected in serum creatinine or in levels of tubular injury biomarkers of AKI. In this proposal, we present new preliminary human data that support these findings. In Aim 1, we will perform the first comprehensive, prospective, nested case-control study of FGF23 and mineral metabolism in 51 patients who developed AKI and 51 controls who remained free of AKI after cardiac surgery. Using preoperative and postoperative samples that were already collected for the Brigham and Women's AKI cohort, we will repeatedly measure intact and C-terminal FGF23, parathyroid hormone, phosphate, calcium, and a complete vitamin D panel, including indices of vitamin D stores (25D), vitamin D activation (1,25D), vitamin D degradation (24,25D) and vitamin D binding protein. Cardiac surgery is an ideal human model of AKI for this Aim because baseline kidney function and the precise timing of renal injury are known, and flanking blood samples are available before and after surgery. Elevated FGF23 levels and associated alterations in other mineral metabolites are also strongly associated with greater risks of cardiovascular disease and mortality in CKD. In contrast, data on the impact of FGF23 levels on clinical outcomes in AKI are sparse. In a pilot study, we reported that FGF23 is elevated in AKI and associated with greater risk of death or need for renal replacement therapy (RRT). In Aim 2, we will test whether elevated FGF23 and disordered mineral metabolism are independent risk factors for adverse outcomes in an efficient case-cohort analysis of 800 participants in the VALID study, which is a large, prospective critical care cohort with stored blood samples and high rates of incident AKI, RRT, and death. By unifying an intensive patient-oriented physiological study and a large clinical outcomes study under one thematic umbrella, we will generate novel, clinically relevant insights into the emerging role of FGF23 in states of kidney dysfunction that could have important diagnostic and therapeutic implications.